The purpose of these studies is to develop imaging techniques to monitor sub-cellular structures and processes, in vivo. The major approach used was non-linear optical microscopy techniques. We have been systematically developing an in vivo optical microscopy system that is adapted to biological tissues and structures rather than forcing an animal on a conventional microscope stage. The following major findings were made over the last year: 1) Minimally invasive, two photon excitation fluorescence microscopy (TPEFM) is being used to study sub-cellular metabolic processes within cells, in intact animals, under normal in vivo conditions using various exogenous and intrinsic fluorescent probes. We have modified our motion tracking schemes to perform full three dimensional motion tracking to compensate for physiological motion in all three dimensions. Using a graphical processing units in a near real time computer and a resonant scanning mirror in the microscope, we able to track tissue motion on the order of 250 msec. This permits the correction of most slow motions inside cells on the micron scale, in vivo. 2) We have expanded our system to permit hybrid two and one photon excitation schemes to permit the use of near UV lasers to photoconvert probes within the tissues in vivo. Our primary target for this adaptation is the tracking mitochondrial motion and fusion events using a photoconverting protein genetically targeted to mitochondria in a transgenic mouse line. 3) We have expanded our efforts in vascular bed structure determination in vivo. Using state of the art image processing approaches we have been able to segment out the vascular structures inside tissues at such a high fidelity that we can predict vascular flow patterns in silica from these data. This is providing new insights into the structure and control of the microcirculation of various tissues, including the segmentation of vascular flow to slow and fast twitch muscle fibers within a single muscle group. 3) We have been continuing our studies on the use of adaptive optics to correct for the distortion of the excitation light in these studies. New methods of using the point spread function of the system as well as iterative image analysis approaches are beginning to make significant first order corrections to the images that we hope will improve image depth, resolution and power requirements in the next reporting period. 4) We have been working with a commercial partner on the development of a total emission detection system to improve the signal to noise and efficiency of TPEFM. This system, as discussed in previous reports, collects all of the scattered light emitting from a tissue to create the image rather than using the light collected by the microscope objective. This results in an improvement of a factor of 3 or 4 in signal to noise from conventional detection systems. The commercial prototype is being installed in the laboratory this fall.